© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 12-TRANSCODER-HP
    7. 22035
    Prev
    Next
    INDEX
    Explanations

    so

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    å®
    -0.28
    artisan
    -0.28
    çĥĤ
    -0.27
    uju
    -0.25
    é£İåij³
    -0.25
    leitung
    -0.25
    oppins
    -0.25
    ulk
    -0.24
    .ResponseWriter
    -0.24
    Charsets
    -0.24
    POSITIVE LOGITS
    wc
    0.29
    istra
    0.26
    obe
    0.25
    çļĦçĶŁæ´»
    0.25
    orges
    0.25
    æį·
    0.25
    prs
    0.24
    lrt
    0.24
    æĵ¾
    0.24
     |/
    0.24
    Activations Density 0.257%

    No Known Activations